Commit graph

8 commits

Author SHA1 Message Date
Solomon Peachy
8cb555460f [3/4] Completely remove HWCODEC support
'swcodec' is now always set (and recording_swcodec for recording-capable
units) in feature.txt so the manual and language strings don't need to
all be fixed up.

Change-Id: Ib2c9d5d157af8d33653e2d4b4a12881b9aa6ddb0
2020-07-24 21:20:13 +00:00
Michael Sevakis
5b08f1a5b9 Remove I/O priority. It is harmful when used with the new file code.
HAVE_IO_PRIORITY was defined for native targets with dircache.

It is already effectively disabled for the most part since dircache no
longer lowers its thread's I/O priority. It existed primarily for the
aforementioned configuration.

Change-Id: Ia04935305397ba14df34647c8ea29c2acaea92aa
2014-08-30 14:01:21 -04:00
Michael Sevakis
6ed00870ab Base scheduler queues off linked lists and do cleanup/consolidation
Abstracts threading from itself a bit, changes the way its queues are
handled and does type hiding for that as well.

Do alot here due to already required major brain surgery.

Threads may now be on a run queue and a wait queue simultaneously so
that the expired timer only has to wake the thread but not remove it
from the wait queue which simplifies the implicit wake handling.

List formats change for wait queues-- doubly-linked, not circular.
Timeout queue is now singly-linked. The run queue is still circular
as before.

Adds a better thread slot allocator that may keep the slot marked as
used regardless of the thread state. Assists in dumping special tasks
that switch_thread was tasked to perform (blocking tasks).

Deletes alot of code yet surprisingly, gets larger than expected.
Well, I'm not not minding that for the time being-- omlettes and break
a few eggs and all that.

Change-Id: I0834d7bb16b2aecb2f63b58886eeda6ae4f29d59
2014-08-16 05:15:37 -04:00
Michael Sevakis
981d028c09 Do some kernel cleanup
* Seal away private thread and kernel definitions and declarations
into the internal headers in order to better hide internal structure.

* Add a thread-common.c file that keeps shared functions together.
List functions aren't messed with since that's about to be changed to
different ones.

* It is necessary to modify some ARM/PP stuff since GCC was complaining
about constant pool distance and I would rather not force dump it. Just
bl the cache calls in the startup and exit code and let it use veneers
if it must.

* Clean up redundant #includes in relevant areas and reorganize them.

* Expunge useless and dangerous stuff like remove_thread().

Change-Id: I6e22932fad61a9fac30fd1363c071074ee7ab382
2014-08-08 01:59:59 -04:00
Michael Sevakis
81ffd9bfee Fix some stuff for no priority and
thread_queue_wake() doesn't need the 2nd parameter. The original purpose
for it never came to be.

Non priority version mrsw_writer_wakeup_readers was left improperly
finished. Get that back into line.

Change-Id: Ic613a2479f3cc14dc7c761517670eb15178da9f5
2014-08-06 02:27:49 -04:00
Michael Sevakis
533d396761 Add multi-reader, single-writer locks to kernel.
Any number of readers may be in the critical section at a time and writers
are mutually exclusive to all other threads. They are a better choice when
data is rarely modified but often read and multiple threads can safely
access it for reading.

Priority inheritance is fully implemented along with other changes to the
kernel to fully support it on multiowner objects.

This also cleans up priority code in the kernel and updates some associated
structures in existing objects to the cleaner form.

Currently doesn't add the mrsw_lock.[ch] files since they're not yet
needed by anything but the supporting improvements are still useful.

This includes a typed bitarray API (bitarray.h) which is pretty basic
for now.

Change-Id: Idbe43dcd9170358e06d48d00f1c69728ff45b0e3
Reviewed-on: http://gerrit.rockbox.org/801
Reviewed-by: Michael Sevakis <jethead71@rockbox.org>
Tested: Michael Sevakis <jethead71@rockbox.org>
2014-08-06 02:47:47 +02:00
Michael Sevakis
355b5406fa Make errno a thread-local variable.
errno is supposed to be thread local and we do that here in a rather
makeshift way by implementing a function that returns a pointer to
the __errno variable in the thread block.

If more serious TLS is required, perhaps it would be worthwhile
implementing it using the linker and the "__thread" storage attribute.
That's a bit overkill just for this. I'm also not liking what I saw
the compiler producing.

Change-Id: I03bc0bd6a89f6e3d6bae7653284ee01054614f9a
Reviewed-on: http://gerrit.rockbox.org/803
Reviewed-by: Michael Sevakis <jethead71@rockbox.org>
Tested: Michael Sevakis <jethead71@rockbox.org>
2014-05-09 23:14:58 +02:00
Thomas Martitz
382d1861af kernel: Break out kernel primitives into separate files and move to separate dir.
No code changed, just shuffling stuff around. This should make it easier to
build only select parts kernel and use different implementations.

Change-Id: Ie1f00f93008833ce38419d760afd70062c5e22b5
2014-03-03 18:11:57 +01:00
Renamed from firmware/export/thread.h (Browse further)